Transluce finds chatbots improving at suicide-risk detection, but subtle failures remain
Sign in✓ Signed in
A new independent evaluation finds that leading AI models are less likely than earlier versions to explicitly encourage suicide or reinforce delusional beliefs. But the same systems can still miss distress when it appears indirectly, or comply with suicide-related creative-writing and farewell-note requests.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...